This invention relates to the molecules with endonucleolytic activity and enhanced half-lives useful in the site-specific cleavage of RNA. This invention also relates to the control of gene expression through the degradation of mRNA.
Ribozymes are RNA molecules with catalytic activities including the ability to cleave at specific phosphodiester linkages in RNA molecules to which they have hybridized, such as mRNAs, RNA-containing substrates, and ribozymes, themselves.
Ribozymes may assume one of several physical structures, one of which is called a "hammerhead." A hammerhead ribozyme is composed of a catalytic core containing nine conserved bases, a double-stranded stem and loop structure (stem-loop II), and two regions complementary to the target RNA flanking regions the catalytic core. The flanking regions enable the ribozyme to bind to the target RNA specifically by forming double-stranded stems I and III. Cleavage occurs in cis (i.e., cleavage of the same RNA molecule that contains the hammerhead motif) or in trans (cleavage of an RNA substrate other than that containing the ribozyme) next to a specific ribonucleotide triplet by a transesterification reaction from a 3', 5'-phosphate diester to a 2', 3'-cyclic phosphate diester. It is currently believed that this catalytic activity requires the presence of specific, highly conserved sequences in the catalytic region of the ribozyme.
Although the endonucleolytic activity of ribozymes has been demonstrated in vitro, their use in vivo has been limited by their susceptibility to RNAses. Furthermore, the production of therapeutics such as ribozymes having greater than 30 or more nucleotides are more expensive and difficult to produce in great quantities. Thus, there is a need for smaller molecules with increased nuclease resistance that can be used to cleave RNA in vitro and in vivo, and to control gene expression for in vivo use.
In an effort to protect antisense oligonucleotides from degradative influences in vivo, various structural modifications have been made to these molecules, including the replacement of phosphodiester linkages with non-phosphodiester linkages, the substitution of various sugar groups and bases, the addition of end-capping groups, and the substitution or replacement of existing structures with the self-hybridizing termini (reviewed in Goodchild (1990) Bioconjugate Chem. 1:165-187; Agrawal et al. (1992) Trends Biotechnol. 10:152-158; WO 93/15194; WO 94/10301; WO 94/12633).
However, modifications which protect an RNA molecule from endonuclease digestion may also affect the catalytic activity of the ribozyme. For example, Perreault et al. (Nature (1990) 344:565-567) report that the replacement of ribonucleotides at various conserved positions within the ribozyme sequence with 2'-deoxynucleotides resulted in a 96% decrease of catalytic efficiency. Perreault et al. (Biochem. (1991) 30:4020-4025) and Dahn et al. (Biochem. (1990) 72:819-23) disclose that the replacement of various 2'-hydroxyl groups with hydrogen atoms reduced the catalytic activity of hammerhead ribozymes. Olsen et al. (Biochem. (1991) 30:9735-9741) report that replacing 2'-hydroxyl groups on all adenosine residues by either fluorine or hydrogen decreases the catalytic activity of a ribozyme. Odai et al. (FEBS Lett. (1990) 267:150-152) report that replacing the exocyclic amino group of a conserved guanosine residue in the core region with hydrogen reduced catalytic activity. Ruffner et al. (Nucleic Acids Res. (1990) 18:6025-6029) and Buzayan et al. (Nucleic. Acids Res. (1990) 18:4447-4451) disclose that replacing oxygen atoms by sulfur on various internucleotide phosphate residues reduces catalytic activity. Pieken et al. (Science (1991) 253:314-317) disclose that catalytic activity is reduced when various 2'-hydroxyl groups on adenosine residues are replaced with fluorine and when the 2'-hydroxyl groups on cytidine residues is replaced with amine groups.
Other groups have substituted nucleotides within the ribozyme with nucleotide analogs. For example, Usman et al. (WO 93/15187) designed chimeric polymers or "nucleozymes" with ribozyme-like catalytic activity having ribonucleotides or nucleic acid analogs (with modified sugar, phosphate, or base) at catalytically critical sites and nucleic acid analogs or deoxyribonucleotides at non-catalytically critical sites.
Recently, modifications such as a reduction in the length of the stem-loop II structure of the hammerhead ribozyme have been made in an effort to design a more stable molecule without reducing its catalytic activity. For example, Goodchild et al. (Arch. Biochem. Biophys. (1991) 284:386-391) replaced stem II and loop II with shorter nucleotide sequences. Tuschl et al. (Proc. Natl. Acad. Sci. (USA) (1993) 90:6991-6994) prepared hammerhead ribozymes with the stem II shortened to two base pairs, closed by a four base-pair loop. McCall et al. (Proc. Natl. Acad. Sci. (USA) 89:5710-5714) replaced the stem-loop with a few nucleotides that cannot form Watson-Crick base pairs between themselves, and/or substituted the stem-loop II and flanking arms with DNA, without reducing activity.
Modifications in ribozyme structure have also included the substitution or replacement of various non-core portions of the molecule with non-nucleotidic molecules. For example, Benseler et al. (J. Am. Chem. Soc. (1993) 115:8483-8484) disclosed hammerhead-like molecules in which two of the base pairs of stem II, and all four of the nucleotides of loop II were replaced with non-nucleoside linkers based on hexaethylene glycol, propanediol, bis(triethylene glycol) phosphate, tris(propanediol)bisphosphate, or bis(propanediol) phosphate. Ma et al. (Biochem. (1993) 32:1751-1758; Nucleic Acids Res. (1993) 21:2585-2589) replaced the six nucleotide loop of the TAR ribozyme hairpin with non-nucleotidic, ethylene glycol-related linkers. Thomson et al. (Nucleic Acids Res. (1993) 21:5600-5603) replaced loop II with linear, non-nucleotidic linkers of 13, 17, and 19 atoms in length.
However, nucleotides in neither the stem-loop region nor nonconserved regions of the catalytic core of the ribozyme have heretofore been replaced with more rigid, non-nucleotidic molecules. Non-nucleotide linkers or inserts with restricted rotation would give a less flexible molecule more in keeping with the requirements of an enzyme.
Thus, what is needed are molecules with improved nuclease resistance and endonucleolytic activity, and which can be quickly and cost-effectively prepared.